AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Modular Robot Design Synthesis with Deep Reinforcement Learning

Julian Whitman, Raunaq Bhirangi, Matthew Travers, Howie Choset

AAAI 2020 · Learning theory and algorithms › Combinatorial and search optimization · approach: Classical and HTN planning · 42 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace IK-based heuristic best-first search with DQN-guided best-first searchmodular design search · Modular manipulator design, 11 modules, 50 target setsavg runtime to first feasible (min): 3.04 -> 0.26Improved
replace genetic algorithm with DQN-guided searchmodular design search · Modular manipulator design, 11 modules, 50 target setsavg complete robot evaluations to first: 138.12 -> 10.31Improved
replace genetic algorithm with DQN-guided searchmodular design search · Modular manipulator design, 11 modules, 50 target setsavg best cost after five minutes: 0.53 -> 0.52Mixed
replace genetic algorithm with DQN-guided search under torque constraintsmodular design search · Modular manipulator design, 16 modules with torque constraint, 50 target setstrials with no feasible robot: 28/50 -> 1/50Improved
replace best-first search with DQN searchmodular design search · Modular manipulator design, 11 modules, 50 target setstrials with no feasible robot: 16/50 -> 1/50Improved
condition the heuristic on obstacle locations (ablation)DQN heuristic · modular design searchsolution quality and runtimeImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy